A stochastic framework to simulate the process of initiation, diameter growth, death and self-pruning of branches in loblolly pine (Pinus taeda L.) trees was developed. A data set was obtained from a destructive sampling of whorl sections from 34 trees growing under different initial spacing. Data from dissected branches were used to develop a model for representing knot shape, which assumed that the live portion of a knot can be modeled by a one-parameter equation and the dead portion by assuming a cylindrical shape. For the developed knot model analytical expressions were derived for estimating the volume of knots (live/dead portions) for three types of branch conditions on simulated trees: (i) live branches, (ii) non-occluded dead branches, and (iii) occluded dead branches. This model was intended to recover information on knots shape and volume during the simulation process of branch dynamics. Three different components were modeled and hierarchically connected: whorl, branches and knots. For each new growing season, whorls and branches are assigned stochastically along and around the stem. Thereafter, branch diameter growth is predicted as function of relative location within the live crown and stem growth. Using a taper equation, the spatial location (X,Y,Z) of both live and dead portion of simulated knots is maintained in order to create a 3D representation of the internal stem structure. At the end of the projection period information on (i) vertical trend of branch diameter and location along and around the stem, (ii) volume of knots, and (iii) spatial location, size and type (live and dead) of knots can be obtained. The proposed branch model was linked to the individual-tree growth and yield model PTAEDA3.1 to evaluate the effect of initial spacing and thinning intensity on branch growth in sawtimber trees. The use of the dynamic branch model permitted generation of additional information on sawlog quality under different management regimes. The arithmetic mean diameter of the largest four branches, one from each radial quadrant of the log (i.e. Branch Index, BI) and the number of whorls per log were considered as indicators of sawlog quality. The developed framework makes it possible to include additional wood properties in the simulation system, allowing linkage with industrial conversion processes (e.g. sawing simulation). This integrated modeling system should promote further research to obtain necessary data on crown and branch dynamics to validate the overall performance of the proposed branch model and to improve its components. / Ph. D.

This study aimed to evaluate the influence of dendrometric, morphometric and qualitative variables in the annual periodic increment in basal area free of trees and under competition Enterolobium contortisiliquum. (Vell.) Morong. with a view to making decisions about the management of the species. The study was conducted in the state's central region of Rio Grande do Sul, in the city of Santa Maria. Subjects were measured and collected the dendrometric, morphometric and qualitative variables. Step data of the last 5 years were obtained by analysis of growth roller collected using the auger Pressler. To determine the density of the trees in the forest, it was calculated basal area in hectare per sampling point based on Bitterlich methodology. The timbaúva differs between the conditions in their dendrometric and morphometric variables, but individuals under competition vary more among themselves that free individuals of competition. The height of the bole in individuals free competition and the variables scope Index and formal canopy in individuals under competition does not suffer variation with increasing DAP values. The shape of the crown of free competition individuals ranges from tolerable perfect, the perfect cup is seen mostly in people over 40 cm DAP. In people under competition exists a variation of forms of canopy are assigned to most individuals as classification of good cup. The crown density of free individuals and under competition is lower in individuals under 40 cm DAP. Individuals under competition have canopy characteristics, being classified as to their sociological position in dominant and codominant. The better the shape and density of the canopy greater the slope of the land and the classification of sociological position in individuals at greater competition will be the height of the shaft. The crown diameter of individuals is estimated as a function of diameter breast height by simple linear model DC = - b0 + b1 * √dap for both conditions. The height of the bole of free individuals of competition has a growth behavior defined by average of 2.20 m approximately, being characteristic of the species forking near the base in the absence of competition. In individuals under competition the stem height can be explained in terms of diameter at breast height the equation HF=b1*√dap. Under competition, the total height is influenced by the density of the population differing in level to the conditions existing the need for using separate models, estimated by the equation HT=b1*√dap+b2*G(bit) and the free individuals competition for equation HT=b1*√dap. The growth in basal area of E. contortisiliquum can be estimated by analysis of stem growth rings and there are differences in growth between the two conditions. The IPAG the competition of free individuals is explained due to the DAP and individuals under competition is explained in terms of DAP and increase the qualitative proximity of water variable. / Este trabalho teve por objetivo avaliar a influência das variáveis dendrométricas, morfométricas e qualitativas no incremento periódico anual em área basal de árvores livres e sob competição de Enterolobium contortisiliquum. (Vell.) Morong. com vistas à tomada de decisões sobre o manejo da espécie. O estudo foi realizado na região Central do estado do Rio Grande do Sul, na cidade de Santa Maria. Os indivíduos foram mensurados e coletadas as variáveis dendrométricas, morfométricas e qualitativas. Os dados de incremento dos últimos 5 anos foram obtidos pela análise de rolos de incremento, coletados com a utilização do trado de Pressler. Para a determinação da densidade entre as árvores na floresta, foi calculada a área basal no ponto de amostragem por hectare baseado na metodologia de Bitterlich. A timbaúva difere entre as condições em suas variáveis dendrométricas e morfométricas, porém indivíduos sob competição variam mais entre si que os indivíduos livres de competição. A altura do fuste nos indivíduos livres de competição e as variáveis Índice de abrangência e formal de copa nos indivíduos sob competição não sofrem variação com o aumento dos valores de DAP. A forma da copa dos indivíduos livres de competição varia de perfeita a tolerável, a copa perfeita é verificada em sua maioria nos indivíduos acima dos 40 cm de DAP. Nos indivíduos sob competição existe uma variação de formas de copa, sendo atribuído para a maioria dos indivíduos a classificação de forma da copa boa. A densidade de copa dos indivíduos livres e sob competição é menor nos indivíduos abaixo dos 40 cm de DAP. Os indivíduos sob competição possuem características de dossel, sendo classificados quanto a sua posição sociológica em dominantes e codominantes. Quanto melhor for a forma e a densidade da copa, maior for a inclinação do terreno e a classificação da posição sociológica nos indivíduos sob competição maior será a altura do fuste. O diâmetro de copa dos indivíduos é estimado em função do diâmetro a altura do peito pelo modelo linear simples DC= -b0+b1*√dap para ambas condições. A altura do fuste dos indivíduos livres de competição possui um comportamento de crescimento definido por média de 2,20 m aproximadamente, sendo característico da espécie bifurcar-se próximo à base na ausência de competição. Nos indivíduos sob competição a altura do fuste pode ser explicada em função do diâmetro a altura do peito pela equação HF=b1*√dap. Sob competição, a altura total sofre influência da densidade do povoamento diferindo em nível para as condições existindo a necessidade do uso de modelos distintos, sendo estimada pela equação HT=b1*√dap+b2*G(bit) e para os indivíduos livres de competição pela equação HT=b1*√dap. O crescimento em área basal da E. contortisiliquum pode ser estimado pela análise dos anéis de crescimento tronco e existem diferenças no crescimento entre as duas condições. O IPAg dos indivíduos livres de competição é explicado em função do DAP e nos indivíduos sob competição é explicado em função do DAP e acréscimo da variável qualitativa proximidade da água.

Individual tree detection and modelling forest parameters using Airborne Laser Scanner data (Light Detection and Ranging (LiDAR) is becoming increasingly important for the monitoring and sustainable management of forests. Remote sensing has been a useful tool for individual tree analysis in the past decade, although inadequate spatial resolution from satellites means that only airborne systems have sufficient spatial resolution to conduct individual tree analysis. Moreover, recent advances in airborne LiDAR now provide high horizontal resolution as well as information in the vertical dimension. However, it is challenging to fully exploit and utilize small-footprint LiDAR data for detailed tree analysis. Procedures for forest biomass quantification and forest attributes measurement using LiDAR data have improved at a rapid pace as more robust and sophisticated modelling used to improve the studies. This thesis contains an evaluation of three approaches of utilizing LiDAR data for individual tree forest measurement. The first explores the relationship between LiDAR metrics and field reference to assess the correlation between LiDAR and field data at the individual-tree level. The intention was not to detect trees automatically, but to develop a LiDAR-AGB model based on trees that were mapped in the field so as to evaluate the relationships between LiDAR-type metrics under controlled conditions for the study sites, and field-derived AGB. A non-linear AGB model based on field data and LiDAR data was developed and LiDAR height percentile h80 and crown width measurement (CW) was found to best fit the data as evidenced by and Adj-R2 value of 0.63, the root mean squared error of the model of 14.8% and analysis of the residuals. This paper provides the foundation for a predictive LiDAR-AGB model at tree level over two study sites, Pasoh Forest Reserve and FRIM Forest Reserve. The second part of the thesis then takes this AGB-LiDAR relationship and combines it with individual tree crown delineation. This chapter shows the contribution of performing an automatic individual tree crown delineation over the wider forest areas. The individual tree crown delineation is composed of a five-step framework, which is unique in its automated determination of dominant crown sizes in a forest area and its adaption of the LiDAR-AGB model developed for the purpose of validation the method. This framework correctly delineated 84% and 88% of the tree crowns in the two forest study areas which is mostly dominated with lowland dipterocarp trees. Thirdly, parametric and non-parametric modelling approaches are proposed for modelling forest structural attributes. Selected modelling methods are compared for predicting 4 forest attributes, volume (V), basal area (BA), height (Ht) and aboveground biomass (AGB) at the species level. The AGB modelling in this paper is extracted using the LiDAR derived variables from the automated individual tree crown delineation, in contrast to the earlier AGB modelling where it is derived based on the trees that were mapped in the field. The selected non-parametric method included, k-nearest neighbour (k-NN) imputation methods: Most Similar Neighbour (MSN) and Gradient Nearest Neighbour (GNN), Random Forest (RF) and parametric approach: Ordinary Least Square (OLS) regression. To compare and evaluate these approaches a scaled root mean squared error (RMSE) between observed and predicted forest attribute sampled from both forest site was computed. The best method varied according to response variable and performance measure. OLS regression was to found to be the best performance method overall evidenced by RMSE after cross validation for BA (1.40 m2), V (1.03 m3), Ht (2.22 m) and AGB (96 Kg/tree) respectively, showed its applicability to wider conditions, while RF produced best overall results among the non-parametric methods tested. This thesis concludes with a discussion of the potential of LiDAR data as an independent source of important forest inventory data source when combined with appropriate designed sample plots in the field, and with appropriate modelling tools.

No description available.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / This work aims at modeling the diameter growth for individual trees of Araucaria angustifolia, present in the Mixed Ombrophylous Forest at National Forest of São Francisco de Paula, northeast of Rio Grande do Sul. Therefore, 8 years remeasurements of 251 araucaria individuals were used, they were distributed over the permanent plots of the Project PELD CNPq. After the data consistency confirmation, a yield model was built, having as dependent variable the future diameter, using regression and analysis of covariance. As independent variable, the dimensional characteristic of the individuals were used and the distance independent competition indices. The most correlated variable with the future diameter was the current diameter. The competition indices showed high correlation with the dependent variable, correctly representing the competitive pressure exerted on the tree, but they didn´t enter in the regression model. Only the current diameter, multiplied by a coefficient, plus a constant value which differed according to the sociological position of the individuals, it was effective to predict the future diameter. The model showed better estimates for emerging individuals of Araucaria angustifolia, however, its use in the individual tree growth prognosis should be cautious due to the magnitude of the errors it presented. / Este trabalho objetivou modelar o crescimento em diâmetro para árvores individuais de Araucaria angustifolia, ocorrentes na Floresta Ombrófila Mista da Floresta Nacional de São Francisco de Paula, no nordeste do Rio Grande do Sul. Para tanto, foram utilizadas remedições de 8 anos de 251 indivíduos de araucária, distribuídos ao longo das parcelas permanentes do Projeto PELD CNPq. Após confirmação da consistência dos dados, foi construído um modelo de produção, tendo como variável dependente o diâmetro futuro, utilizando regressão e análise de covariância. Como variáveis independentes, foram utilizadas as características dimensionais dos indivíduos e os índices de concorrência independentes de distância. A variável mais correlacionada com o diâmetro futuro foi o diâmetro atual. Os índices de concorrência apresentaram alta correlação com a variável dependente, representando corretamente a pressão competitiva exercida sobre a árvore, mas não entraram no modelo de regressão. Somente o diâmetro atual, multiplicado por um coeficiente, acrescido de um valor constante o qual diferiu de acordo com a posição sociológica do indivíduo, foi eficiente para predizer o diâmetro futuro. O modelo apresentou melhores estimativas para indivíduos de Araucaria angustifolia emergentes, no entanto, o seu emprego na prognose de crescimento de árvores individuais deve ser cauteloso devido a magnitude dos erros de predição que apresentou.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Reliable growth data from trees are very important in the context of forest management. Tree-characteristics such as architecture and competition are associated with diameter growth. Previous studies analyzing tree growth changes have indicated that these variables can describe these effects more effectively. However, the precise role of these effects in the growth modeling of tropical trees needs to be studied further to discuss more effectively the actual forest management practices. Here, we reconstructed the periodic annual basal area increment (IPAg) of four long-lived timber species from the Amazon forest: Cedro (Cedrela odorata); Cerejeira (Amburana cearensis); Copaíba (Copaifera paupera) e Mogno (Swietenia macrophylla) aiming to explain the IPAg variation by developing a individua-tree growth model using the following potential predictors: tree size, morphometric data, competition; crown position and liana load. The model was fitted using a Gamma distribution in the context of Generalized Linear Models. The statistics criteria and residual distributions showed that growth model was adequate to adjust the observed IPAg data and revealed some dendrometric variables that were involved into the effects that lead the observed variation in IPAg, presenting a broad range of the observed data in situations were growth is complex and variable between species. Trees sampled in good site conditions - expressed as a good crown exposure to light, and therefore low competition - presented higher growth rates than trees sampled under high competition from neighboring trees. Besides, the variables Liana load, for Cedrela, the crown diameter and the Hegyi competition index was demonstrated to be highly influential predictive variable. Also, trees sampled in densities greater than 25 m2.ha-1 had their IPAg% significantly reduced. In the context of basal area increment yield by the space occupied in the forest (crown efficiency) the species presented mostly efficiency when growth in dominant position with thin and long crown (crown formal). Among the species, Cedrela odorata was the most efficient in growth. This behavior reinforces the importance of release trees to obtain better crown exposure. / Dados confiáveis de crescimento de árvores são muito importantes no contexto do manejo florestal. Características próprias da árvore como a arquitetura e a competição estão associadas diretamente com o crescimento em diâmetro e altura. Estudos anteriores indicaram que essas variáveis podem descrever a variação do incremento. No entanto, o efeito dessas variáveis na modelagem do crescimento de árvores tropicais precisa ser mais bem estudado, para discutir efetivamente as praticas de manejo florestal atuais. Portanto, nessa pesquisa reconstruiu-se o incremento periódico anual em área basal (IPAg) de quatro espécies de árvores madeiráveis de vida longa da Amazônia: Cedro (Cedrela odorata); Cerejeira (Amburana cearensis); Copaíba (Copaifera paupera) e Mogno (Swietenia macrophylla) com o objetivo de explicar a variação de IPAg, propondo um modelo de crescimento tipo de árvore individual utilizando os seguintes preditores potenciais: tamanho e morfometria da árvore; o status competitivo; a posição social e a carga de lianas na copa. O modelo foi ajustado considerando a distribuição de probabilidade Gama no contexto de Modelo Linear Generalizado. Em base a critérios estatísticos e avaliação residual, o modelo de crescimento demonstrou ser adequado para explicar a variação de IPAg e revelou variáveis dendrométricas que englobam grande parte dos efeitos que moldaram a variação observada do IPAg, apresentando boa abrangência dos dados observados em situações onde o crescimento é complexo e variado entre as espécies. As árvores quando amostradas em boas condições de sítio, expressado por boa exposição da copa à luz solar e, portanto, pouca competição, apresentaram as maiores taxas de IPAg, no período avaliado, do que aquelas encontradas sob forte competição por árvores vizinhas. Além disso, as variáveis carga de lianas na copa, para Cedrela, o diâmetro de copa e o status competitivo de Hegyi influenciaram de forma significativa o IPAg. Árvores amostradas em densidade maior do que 25 m2.ha-1 mostraram queda significativa do IPAg%. No contexto de incremento em área basal por área ocupada na floresta (eficiência da copa), todas as quatro espécies mostraram-se mais eficientes quando em melhor posição social e quando com copas estreitas e longas (formal de copa pequeno). Entre as espécies a Cedrela odorata foi a mais eficiente no crescimento, comportamento que reforça a importância da liberação das árvores o que proporcionaria melhor exposição da copa à luz solar.

Data de aprova??o retirada da vers?o impressa do trabalho. / Submitted by Jos? Henrique Henrique (jose.neves@ufvjm.edu.br) on 2017-09-05T18:36:05Z No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) tamires_mousslech_andrade_penido.pdf: 7601696 bytes, checksum: 5db79c506834a030e0682976f71d2692 (MD5) / Approved for entry into archive by Rodrigo Martins Cruz (rodrigo.cruz@ufvjm.edu.br) on 2017-09-18T17:02:44Z (GMT) No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) tamires_mousslech_andrade_penido.pdf: 7601696 bytes, checksum: 5db79c506834a030e0682976f71d2692 (MD5) / Made available in DSpace on 2017-09-18T17:02:44Z (GMT). No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) tamires_mousslech_andrade_penido.pdf: 7601696 bytes, checksum: 5db79c506834a030e0682976f71d2692 (MD5) Previous issue date: 2017 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior (CAPES) / A modelagem ? um procedimento estat?stico empregado por gestores florestais para esbo?ar o desenvolvimento vegetal com precis?o. Informa??es confi?veis do crescimento e da produ??o s?o essenciais para predizer e quantificar a estrutura futura do povoamento. O presente trabalho foi dividido em dois cap?tulos. Os objetivos foram avaliar a efici?ncia de se estimar a altura empregando diferentes modelos hipsom?tricos, crit?rios de estratifica??o e m?todos de ajuste, al?m de comparar tr?s categorias de modelos de crescimento e produ??o (MCP) em planta??es comerciais de eucalipto. Foram definidas quatro unidades de manejo florestal, totalizando 293,43 ha. O invent?rio florestal cont?nuo foi realizado em 34 parcelas permanentes de 400 m2. O espa?amento de plantio foi de 3,0 x 2,5 m. Avaliou-se a precis?o do ajuste de treze modelos hipsom?tricos. Foram treinadas RNA empregando as mesmas vari?veis de resposta e preditoras adotadas nas equa??es selecionadas. As categorias de MCP testadas foram: em n?vel de povoamento (MP), pelo sistema de equa??es simult?neas de Clutter; de distribui??o diam?trica (MDD), pelo ajuste de fun??o densidade de probabilidade de Weibul-2P e de ?rvores individuais (MAI), pelo modelo de Pienaar e Schiver. As equa??es provenientes do modelo de altura em fun??o do di?metro e da altura dominante forneceram estimativas confi?veis da altura para diferentes crit?rios de estratifica??o, demonstrando superioridade em rela??o aos modelos locais. A modelagem por regress?o e redes demonstraram-se adequadas para estimar a altura, com ou sem estratifica??o do banco de dados. A estratifica??o ? um procedimento que pode melhorar a qualidade das estimativas de altura obtidas por regress?o e RNA. As tr?s categorias de modelo proporcionaram estimativas confi?veis da produ??o em volume com casca, aos 36, 48, 60 e 72 meses, para as unidades de manejo estudadas. O MAI foi a categoria mais precisa e consistente na estimativa do volume por hectare. As proje??es com MP e MDD podem gerar estimativas similares de volume para idades al?m daquelas em que se realizou o invent?rio florestal. / Disserta??o (Mestrado) ? Programa de P?s-Gradua??o em Ci?ncia Florestal, Universidade Federal dos Vales do Jequitinhonha e Mucuri, 2017. / Modeling is a statistical procedure employed by forest managers to sketch plant development with precision. Reliable growth and production information are essential to predict and quantify the future stand structure. The present work was divided in two chapters. The objectives were to evaluate the efficiency of height estimation using different hypsometric models, stratification criteria and adjustment methods, beside to evaluate and compare three categories of growth and yield models (MCP) in commercial eucalypt plantations. Four forest management units were defined, totaling 293.43 ha. The continuous forest inventory was realized in 34 permanent plots of 400 m2. The planting spacing was 3.0 x 2.5 m. The accuracy of the fit of thirteen hypsometric models was evaluated. ANN were trained using the same response e predictive variables adopted in the selected equations. The MCP categories tested were: in level of stand (MP), using Clutter?s simultaneous equations; diameter distribution model (MDD), by adjustment of the Weibull-2P?s probability density function and individual trees (MAI), by Pienaar and Schiver model. The equations from the height model according to the diameter and the dominant height provided reliable height estimates for different stratification criteria, showing superiority in relation to local models. Regression and networks modelling were suitable for estimating height, with or without stratification of the database. Stratification is a procedure that can improve the quality of the estimates obtained by regression and ANN. The three model categories provided reliable estimates of the volume with bark production at 36, 48, 60 and 72 months for the management units studied. MAI was the most accurate and consistent category in estimating volume per hectare. Projections with MP and MDD can generate similar estimates of volume for ages beyond those in which the forest inventory was carried out.

Made available in DSpace on 2015-03-26T12:27:15Z (GMT). No. of bitstreams: 1 texto completo.pdf: 8085749 bytes, checksum: 16c9ec262ce793f99b1a887855d560a4 (MD5) Previous issue date: 2012-12-21 / Conselho Nacional de Desenvolvimento Científico e Tecnológico / This study aimed to develop a complete model in individual tree basis for a fragment of Semideciduous Montana forest using artificial neural networks (ANN) and validate their application. Data from ten permanent plots in a forest of 17 ha in Viçosa, MG, at the intermediate stage of ecological succession, monitored for 14 years (1994, 1997, 2000, 2004 and 2008) were used for this purpose. The data set was randomly divided into two groups: 1) training networks group, consisting of six parcels totaling 3,556 cases at five measurements, observed 231 cases of death and 238 cases of ingrowth, and 2) the validation model group, consisting of four parcels, totaling 2,062 cases, with 181 cases of death and 146 cases of ingrowth. The work was divided into four chapters where, in Chapter I, we evaluated different competition indexes independent, dependent and semi-independent of distance, and selected those best for studies of growth and mortality in later chapters. In Chapters II and III were evaluated and compared different types and architectures of neural networks to estimate mortality regular trees, as well as projection of diameter, total height and bole height. In Chapter IV, there was the validation of networks employing the set of independent data. Besides the selected networks was considered the average rate at each measurement period, the first data group, the projection of the forest structure (validation). The analysis showed that, overall, the competition indexes were correlated with the probability of mortality and growth of trees. There was also verified that the neural networks efficiency in the estimation of tree mortality and the growth projection, accurate estimates are obtained. In validation, designed to mortality, growth and ingrowth of the first measurement (1994) by the year 2008, which were compared to the observed values. The model for individual tree basis was efficient in simulating forest growth and yield at all levels of detail (stand level, diameter and hight distribution level and individual tree level). The results confirmed the potential use of individual tree basis models through artificial neural networks in natural forests structurally complex, as is the case of the Semideciduous Montana Forest. / Este estudo teve como objetivos desenvolver um modelo completo em nível de árvore individual para um fragmento de Floresta Estacional Semidecidual Montana utilizando redes neurais artificiais (RNA) e validar a sua aplicação. Dados provenientes de dez parcelas permanentes em um fragmento florestal de 17 ha no município de Viçosa, MG, em estágio médio de sucessão ecológica, monitoradas durante 14 anos (1994, 1997, 2000, 2004 e 2008) foram utilizados para este propósito. O conjunto de dados foi dividido aleatoriamente em dois grupos: 1) grupo de treinamento das redes, composto por seis parcelas e totalizando 3.556 casos nas cinco medições, sendo observados 231 casos de mortalidade e 238 casos de ingresso, e 2) grupo de validação do modelo, composto por quatro parcelas, totalizando 2.062 casos, sendo observados 181 casos de mortalidade e 146 casos de ingresso. O trabalho foi dividido em quantro capítulos onde, no capítulo I, foram avaliados diferentes índices de competição independentes, dependentes e semi-independentes da distância, sendo selecionados aqueles melhores para os estudos de crescimento e mortalidade nos capítulos posteriores. Nos capítulos II e III foram avaliados e comparados diferentes arquiteturas e tipos de redes neurais para estimativa da mortalidade regular das árvores, bem como para projeção do diâmetro, altura total e do fuste. No capítulo IV, realizou-se a validação das redes empregando-se o conjunto de dados independentes. Além das redes selecionadas considerou-se a taxa média de ingresso em cada período de medição, no primeiro grupo de dados, na projeção da estrutura da floresta (validação). Após as análises, observou-se que, de maneira geral, os índices de competição foram correlacionados com o crescimento e probabilidade de mortalidade das árvores. Verificou-se, também, a eficiência das redes neurais na estimação da mortalidade das árvores e para a projeção do crescimento, sendo obtidas estimativas precisas. Na validação, projetou-se a mortalidade, o crescimento e o ingresso da primeira medição (1994) até o ano de 2008, os quais foram comparados aos valores observados. O modelo em nível de árvore individual foi eficiente na simulação do crescimento e produção florestal em todos os níveis de detalhamento (povoamento total, distribuição diamétrica e de altura e em nível de árvore individual). Os resultados confirmaram o potencial de utilização de modelos em nível de árvore individual por meio de redes neurais artificiais em florestas naturais estruturalmente complexas, como é o caso da Floresta Estacional Semidecidual Montana.

Segmentation of an image into individual tree crowns is a key step in the processing of remotely sensed data for forestry practice. The doctoral thesis gives a broad overview of this topic. It comprehends theoretical context from mathematical point of view and defines basic terms from airborne imaging and laser scanning. Mathematical methods of tree detection are focused on a robust adaptation to the actual conditions in a region of interest. A novel approach of crown area delineation is introduced, it combines a seeded region growing technique with an active contour as a crown boundary representation. The parametrisation of all algorithms is analysed in a practical half of the thesis and more application-oriented issues are mentioned. Executable computer programs are attached.

No description available.